Abstract
Aluminum phosphates are materials with relatively wide potential applications in many industries. The vibrational features of selected compounds were established on Raman and infrared spectroscopy. The experimentally determined spectra are compared to those calculated by ab initio methods. This gives a unique possibility of a proper assignment of the experimental spectral features to specific modes of vibration. In the results, it was evidenced that the spectra are characterized by two specific intense bands in the mid- and high-frequency range due to the P–O–P and P–O bonds in [PO4] tetrahedron vibrations. The position of the high-frequency band is related to the number of bridging oxygen atoms connecting [PO4] tetrahedrons in the unit cell. Additionally, the differences in the spectra were evidenced as a result of different polymorphic forms of the selected compounds. Therefore, the results may be useful in determining the phase composition of polyphase materials or structural features of aluminum–phosphate glasses and glass–ceramic materials.
Similar content being viewed by others
Introduction
Aluminum phosphates are present in many applications, for example in chemically bonded phosphate ceramics (CBPC) with alumina, dental cement, refractory binders, composite materials, and glass–ceramics1,2,3,4,5,6,7,8,9,10. Pyrophosphates containing aluminum and monovalent cations, such as NaAlP2O7, can be used as solid electrolytes for batteries, piezoelectric and ionic conductors11,12,13. Furthermore, NaAlP2O7 with different doped rare earth ions has potential application in white light emitting diodes (WLEDS)14,15. In the group of aluminum phosphates are molecular sieves (AlPO) that can be used in catalysis, separation, and ion exchange16,17,18.
Raman and infrared spectroscopies (IR), in addition to X-ray diffraction (XRD), are one of the most important methods of structural characterization of different materials. The spectroscopies are especially important in the case of amorphous materials such as glasses, where, because of the lack of long-range order, application of XRD is strongly limited. In this method, the proper assignment of characteristic bands to specific vibrations is a crucial point. To solve the problem, calculation methods based on density functional theory (DFT) can be very helpful. The methods allow for the prediction of theoretical IR and Raman spectra with considerable precision19,20,21,22,23.
The aim of the work was to compare theoretical and experimental IR and Raman spectra of different aluminum-phosphate compounds. Additionally, the theoretical results were used to determine the proper assignment of the characteristic spectral features to the different vibration modes. Special attention was paid to the position of the bands related to bond vibrations in the [PO4] tetrahedrons. The structural elements are the main building blocks of the aluminum phosphate compounds. Moreover, it is interesting to observe their changes resulting from structural transformations e.g. from chain to ring structures that may be evidenced in the compounds.
In the work, the Qi notation is applied as commonly used in phosphate glasses19,24. In this notation, ‘Q’ means phosphorus tetrahedron [PO4], and ‘i’ is a number of other phosphorus tetrahedra connected to ‘Q’. Aluminum phosphates were chosen so that all Qi structural units were represented in the studies. Only the structural units Q3 are in pure P2O5, where the most stable polymorphic form is o’-P2O525. Pure Q2 units are characteristic for Al(PO3)3, which has three polymorphs. A-Al(PO3)3 and aluminum cyklohexaphosphate, which have 4- and 6-membered [PO4] rings, respectively. Although B-Al(PO3)3 has a chain structure4,26,27,28. Aluminum cyklohexaphosphate and B-Al(PO3)3 are stable at temperatures lower than 800 °C but detailed studies have not been carried out. Above 800 °C, mainly A-Al(PO3)3 has been reported (Fig. 1)4,28,29,30. There is no pure aluminum–phosphate compound in which there are only Q1 units. Therefore, sodium-containing NaAlP2O7 was chosen. In the crystal structure, P2O74− dimers are present that are two joined Q1 structural units. The case of Q0 is represented by AlPO4, which is one of the most studied compounds of aluminum phosphates4,7,31,32,33,34,35,36,37. AlPO4 is a high refractory material with a melting point of about 1950 °C38 but a glaze on the surface due to probably the loss of P2O5 can be detected32. It undergoes several phase transformations, as shown in Fig. 139, and the phases are isostructural to SiO2.
In the work low-temperature form of berlinite (α-berlinite) which is isostructural to α-quartz and low-temperature α-cristobalite type AlPO4 that has a close similar structure to α-cristobalite31,33 were studied.
Although the selected compounds are known, the number of literature data concerning their vibrational features is relatively limited. To the best of our knowledge, this is the first report in which all of the compounds are gathered together, and their experimental spectra are compared with the theoretical ones.
Results
o’-P2O5
The DFT optimized unit cell of o’-P2O5 is shown in Fig. 2. In the unit cell, there exist only Q3 structural units. In the unit cell, 3 bridging oxygens are involved in the formation of P–OB–P bridges, and one is double-bonded to oxygen P=O. There are two inequivalent phosphorous sites with the mean P–O bond lengths 1.573 Å, 1.446 Å for P–OB and P=O, respectively. The shorter length of the P=O bond leads to distortion of the [PO4] tetrahedron with the off-center shift of the central atom. The calculated Raman and IR intensities and their assignments are summarized in detail in Table S1.1 (supplementary materials). The calculated vibrations for o’-P2O5 were assigned to the vibrations of the idealized Q3 molecule (points group C3v) and idealized P-OB-P bridge (points group C2v). (Fig. S1.1 and S1.2 supplementary material).
Figure 3 shows the calculated Raman and IR spectra, and the simplified frequency ranges of the specific vibrations are summarized in Table 1. As can be seen in the Raman spectrum the most intense bands are at 604 cm−1 and 1300, 1344 cm−1. The lower frequency band is related to the symmetric (A1) and symmetric deformation (E) of 3(P-OB) in Q3. The higher value is due to the stretching of P=O in Q3 units. Other vibrations are considerably weaker. In the case of the IR spectrum, the strongest bands at 937 and 957 cm−1 are related to asymmetric stretching of P–OB–P and 3(P–O) in Q3. In this case, the bands due to P=O vibrations are also present, although their intensities are considerably lower.
It should be noted that in this case, we present only theoretical spectra that were not scaled or shifted.
B-Al(PO3)3
The B-Al(PO3)3 is made up of infinitely twisted chains of structural units connected by [AlO6] octahedra. The unit cell is shown in Fig. 4. The length of the P-OB bond changes in the range of 1.570–1.600 Å, whereas that of P-ONB varies in the range of 1.480–1.488 Å. It should be pointed out that in the crystal structure there are no pure double-bonded oxygen atoms (P=O). All the non-bridging oxygens form P-ONB-Al bridges, and the excess phosphorus positive charge is redistributed over the two P-ONB bonds.
The calculated Raman and IR spectra are shown in Fig. 5. Detailed vibration assignments to idealized Q2 of the C2v point group and their positions are summarized in Table S1.2 (supplementary material) and in simplification in Table 2.
The Raman spectrum of B-Al(PO3)3 is characterized by two intense bands. The strongest one is at 1186 cm-1 and weaker at 640 cm−1. The higher frequency band is related to the symmetric stretching vibrations (A1) of 2(P–ONB) in the Q2 structural units. The second lower frequency is due to symmetric stretching (A1) in P-OB-P and bending (A1) of 2(P-ONB) in Q2.
The IR spectrum of B-A(PO3)3 is more complex. In the range of 922–1350 cm−1, there are three groups of strong bands. The first group between 922 and 1021 cm−1 contains asymmetric stretching vibrations (B1) of 2(P-OB) in Q2 and asymmetric stretching (B1) in P-OB–P. The group between 1047 and 1149 cm−1 is related to the symmetric stretching (A1) of 2(P–ONB) and the symmetric stretching (A1) of 2(P–OB) in Q2. The last group between 1201 and 1350 cm−1 is due to the asymmetric stretching (B1) of 2(P–ONB) in Q2. The medium intensity in the IR spectrum has vibrations related to bending (A1) in P–OB–P and deformations in Q2 in the range of 250–617 cm−1. Other modes are much weaker.
A-Al(PO3)3
The powder XRD diffraction pattern of the synthesized sample containing A-Al(PO3)3 is shown in Fig. 6. The Rietveld refinement of the data showed that in the sample two phases can be distinguished. The main crystalline phase is A-Al(PO3)3 in a quantity of approximately 99 wt% and the minority phase is an α-cristobalite type of AlPO4. The detailed composition of the material is given in Table S2.1 (supplementary data). The A-Al(PO3)3 crystallizes in a cubic I \(\overline{4 }\) 3d space group and the fitted basic crystal structure parameter is a = 13.727(6) Å.
In contrast to B-Al(PO3)3 in the A-Al(PO3)3 phase the phosphate network forms 4-membered rings of Q2 structural units (4Q2 ring), and the rings are connected by polyhedrons [AlO6]. In this case, the length of the P-OB bond is in the range of 1.583–1.595 Å, whereas for P-ONB it is in the range of 1.471–1.479 Å. Similarly, as in B-Al(PO3)3 there are no pure double P=O bonds, and all the non-bridging oxygens take part in the formation of P-ONB-Al bridges.
The calculated and experimental Raman and IR spectra are presented in Fig. 7. The corresponding vibrations are summarized in detail in Table S1.3 (supp.) and shortened in Table 3. The 4Q2 rings have S4 space group symmetry and some characteristic vibrations in A-Al(PO3)3 were assigned to this symmetry. It can be seen that there is a very good agreement between the experimental and theoretical results. In both the case of intensities and positions. However, the theoretical spectra were shifted by a constant value of about + 30 cm−1 for both Raman and IR results.
The Ramana spectrum of A-Al(PO3)3 has a very strong band at around 1235 cm−1 due to the symmetric stretching (A1) of 2(P-ONB) in Q2. The second band of lower intensity at 654 cm−1 is related to symmetric stretching (A1) in P–OB–P and bending (A1) of 2(P-ONB) in Q2. In the case of the studied phase, there exist characteristic ring vibrations as presented in Fig. 8. Two bands related to the vibrations at c.a. 1270 cm−1 and 1295 cm−1, which is due to asymmetrical and symmetrical vibrations about a fourfold inversion axis of the 4Q2 ring, respectively. It should be noted that the vibrations are characteristic for A-Al(PO3)3 and are not present in B-Al(PO3)3. Thus, it can be used to distinguish between the two phases.
The IR spectrum has two groups with strong bands at c.a. 886–1046 cm−1 and 1283–1405 cm−1. The first group is related to the stretching modes 2(P–OB) in Q2 and asymmetric stretching (B1) in P–OB–P. The band group 1283–1405 cm-1 is related to asymmetric stretching vibrations (B1) of 2(P-ONB) in Q2. The bands related to the bending modes (A1) in P–OB–P and deformations of 2(P-ONB)) in Q2 are in the range of 286–651 cm−1 and have a medium intensity. Also, visible in the IR spectra are bands related to symmetric stretching (A1) in P-OB-P and bending (A1) of 2(P-ONB) in Q2 in the range of 654–801 cm−1.
Aluminum cyclohexaphosphate—Al(PO3)3
Another polymorphic form of Al(PO3)3 is aluminum cyclohexaphosphate. The powder X-ray diffraction pattern of the synthesized material is shown in Fig. 9. According to the Rietveld analysis, the assumed phase is the main (c.a. 85 wt%) and the rest is A-Al(PO3)3. The detailed phase composition of the material is summarized in Table S2.2 (supp.). The main phase crystallizes in a monoclinic P121/c1 space group and the fitted crystal structure parameters are a=6.072(2) Å, b = 15.036(1) Å, c = 8.182(9) Å, β = 105.12°.
The crystal structure of the aluminum cyclohexaphosphate Al(PO3)3 is similar to A-Al(PO3)3 built of rings that, on the contrary, are composed of 6Q2 units connected by [AlO6] octahedra. In this case, the length of the P-OB bond is in the range of 1.581–1.598 Å, whereas for P-ONB it is in the range of 1.475–1.488 Å.
The calculated and experimental Raman and IR spectra are presented in Fig. 10. The corresponding vibrations are summarized in detail in Table S1.4 (supp.) and shortened in Table 4. The 6Q2 rings have Ci space-group symmetry and some characteristic vibrations were assigned to this symmetry. Similarly to previously, there is good agreement between the experimental and theoretical results. The best convergence is obtained when the theoretical spectrum is shifted by the constant value of c.a. + 25 cm-1.
The Raman spectrum of aluminum cyclohexaphosphate is similar to those of A-Al(PO3)3 and B-Al(PO3)3. The strongest band is related to the symmetric stretching modes (A1) of 2(P-ONB) in Q2. The position of the band is c.a. 1215 cm−1. The second strong band is at c.a. 715 cm-1 and is due to symmetric stretching vibrations (A1) in P-OB-P and bending modes (A1) of 2(P-ONB) in Q2. There are also characteristic 6Q2-ring modes active like symmetric vibrations Ag (Fig. 11) in the range of 1121–1341 cm−1 and 561 cm−1.
In the IR spectrum, the strongest vibrations are in the range of 883–1082 and 1222–1341 cm−1. The first group is related to the stretching of 2(P–OB) Q2 and the asymmetric stretching (B1) in P–OB–P. The second group is related to the asymmetric stretching (B1) of 2(P-ONB) in Q2. Good visible vibrations in the range 1088–1215 cm-1 are related to the asymmetric Au vibration of the 6Q2-ring (Fig. 11).
NaAlP2O7
In Al2O3-P2O5 there is no known pure compound containing Q1 structural units. Therefore, sodium-containing NaAlP2O7 was chosen where the unit cell is built of Q1–Q1 dimers. The XRD pattern of the synthesized material is presented in Fig. 12. The main crystal phase present in the obtained material is NaAlP2O7 (c.a. 85 wt%). Secondary minor phases are AlPO4 of the berlinite and cristobalite type and Al2O3. The detailed phase composition is given in Table S2.3 (suppl.). The main phase crystallizes in a monoclinic P121/c1 space group and the fitted crystal structure parameters are a = 7.197(4) Å, b = 7.704(5) Å, c = 9.314(5) Å, β = 111.72(5)o.
In the unit cell Q1-Q1 (P2O7) dimers are connected to [AlO6] octahedra and sodium polyhedra. In this case, the distance of P-OB is in the range of 1.612–1.616 Å and P-ONB in the range of 1.499–1.527 Å. The excess of the P positive charge is now redistributed over 3 non-bridging oxygens in the Q1 unit. Because the unit has only one bond longer (P–OB) and three of similar lengths (P-ONB), the idealized symmetry of the unit is the same as the Q3 unit. Thus, Q1 has the same C3v point group as Q3.
The calculated and measured Raman and IR spectra are presented in Fig. 13. Good agreement is also observed between the theory and the experiment. The best results may be obtained after including about + 40 cm−1 shift of the theoretical spectra. The detailed positions and intensity of the calculated active bands are summarized in the supp. (Table S1.5), and in the simplified form in Table 5.
The most intense band in the Raman spectrum is at 1055 cm−1 related to the symmetric stretching (A1) modes of 3(P–ONB) in Q1. With this feature are associated bands of higher frequencies in the range of 1073–1251 cm-1 related to asymmetric stretching modes (E) of 3(P-ONB) in Q1. However, the intensity of the asymmetric vibrations is considerably lower. The second strong band is at 737 cm-1 due to symmetric stretching (A1) in P–OB–P and symmetric deformation (A1) of 3(P–ONB).
The IR spectrum is characterized by two strong groups of bands. The first in the range of 892–921 cm−1 is related to asymmetric stretching vibrations (B1) of P–OB–P. The second in the range of 1073–1251 cm−1 is related to the asymmetric stretching modes (E) of 3(P–ONB) in Q1. The medium strength has bands related to bending (A1) in P–OB–P, asymmetric deformation (E) of 3(P-ONB), and symmetric stretching (A1) of 3(P-ONB) (see Table 5).
α-Cristobalite type AlPO4
The powder XRD diffraction pattern of the cristobalite type of AlPO4 is presented in Fig. 14. In this case, we were unable to obtain the pure phase. The main crystalline compound was assumed to be AlPO4 (c.a. 67 wt%). The rest of the crystalline phases in the sample are Al2O3 (c.a. 16 wt%) with the minor addition of A-Al(PO3)3 and the berlinite type of AlPO4. The detailed phase composition and the quantified analysis are summarized in Table S2.4 (supp.). The main phase crystallizes in a orthorombic C2221 space group and the fitted crystal structure parameters are a = 7.103(4) Å, b = 7.096(3) Å, c = 7.011(5) Å.
The α-cristobalite type AlPO4 is built of Q0 structural units connected by [AlO4] tetrahedra. In the crystal structure, there are no bridging oxygen atoms, and all the oxygens are non-bridging. The length of the P-ONB bond is in the range of 1.521–1.523 Å. Due to the fact that all oxygens in the [PO4] tetrahedrons have a similar P-O distance, the tetrahedron is close to ideal and can be described by symmetry of the Td point group.
The calculated and experimental Raman and IR spectra of the samples are presented in Fig. 15. The material obtained is polyphase in the case where characteristic vibrations of A-Al(PO3)3 were also detected in the IR spectra. On the other hand, Raman spectroscopy is measured at a point, and it was possible to detect the spectrum of the pure cristobalite phase. In this case, good agreement between theory and experiment can also be evidenced. The best results were obtained when the calculated spectra had been shifted to a value of + 20 cm−1. A detailed description of the active modes is given in Table S1.6 (supp.) and the simplified version in Table 6.
The Raman spectrum is characterized by a strong band at 1109 cm−1 due to symmetric stretching vibrations (A1) in Q0. In the spectrum there are also visible 3 characteristic bands in the range of 239–735 cm-1. The two in the range of 325–735 cm−1 are related to the symmetric bending (E) and asymmetric deformation (F2) modes of Q0. The band 239 cm−1 is due to lattice vibrations.
The most characteristic feature of the IR spectrum is a strong band at 1106 cm−1 that may be assigned to asymmetric stretching modes (F2) in Q0. Also, in the IR spectra there are good visible medium vibrations related to symmetric bending (E) of Q0 and weak vibrations related to asymmetric deformation (F2) of Q0.
α-Berlinite AlPO4
The next polymorphic form of AlPO4 is the berlinite type. The XRD pattern of the synthesized material is given in Fig. 16. As can be seen, the synthesized material was polyphase. The main crystalline compound is the assumed berlinite type of AlPO4 (c.a. 57 wt%). There are also Al2O3, A-Al(PO3)3, and other polymorphic phase of AlPO4 such as cristobalite. The detailed phase composition of the synthesized sample is summarized in Table S2.5 (supp.). The main phase crystallizes in a trigonal P3221 space group and the fitted crystal structure parameters are a = b = 4.948(5) Å, c = 10.950(7) Å.
The crystal structure of berlinite is very similar to that of cristobalite AlPO4. The unit cell is composed of Q0 structural units connected by [AlO4] tetrahedra with the P-OB distance in the range of 1.507–1.512 Å.
The calculated and experimental Raman and IR spectra are shown in Fig. 17. In this case, the best agreement is obtained for the theoretical data shifted by + 25 cm−1. Similarly to the above, the most suitable Raman spectrum was chosen to compare with the calculated spectra. In the IR spectrum, in addition to bands related to α-cristobalite type AlPO4 there are weak bands of A-Al(PO3)3. The detailed Raman and IR active modes with the proper assignment are given in Table S1.7 (supp.) and the simplified in Table 7.
On the Raman spectrum, the strongest band at 1109 cm−1 may be assigned to vibrations of symmetric stretching (A1) in Q0. The position of this band is very similar in α-cristobalite type AlPO4 and α-berlinite. The medium bands are present in the range of 451–741 cm−1 and are related to the symmetric bending modes (E) of Q0. The bands related to the asymmetric deformation (F2) of Q0 are very weak in the α-berlinite spectrum.
On the IR spectrum, the strongest band is at 1092 cm−1 due to asymmetric stretching vibrations (F2) in Q0. Furthermore, bands in the range of 451–741 cm−1 related to asymmetric deformation (F2) of Q0 are clearly visible in the spectrum.
Discussion
Analyzing the obtained experimental and theoretical results one may see that the main intense bands are due to P-O bonds vibrations in Qi structural units in the higher frequency range and P-O-P in the midregion. This is the most well seen in the case of the Raman spectra, wherein in the most considered cases the two bands are dominating. The intensity of the midband decreases with the Qi index, which is related to the decrease of the number of P–O–P linkages. The position of the symmetric modes is centered at frequencies lower than asymmetric.
The position of the bands related to Qi units depends on the value of the parameter i, and with the parameter increase the position shifts towards higher values, which is presented in Fig. 18. The separated ranges of the vibrations for specific Qi spices can be distinguished. This shows that Raman and IR spectroscopies may provide important information concerning the Qi distribution in materials.
On the other hand, the modes related to the different vibrations in [AlOx] polyhedrons are very weak, and it seems that spectroscopies cannot be utilized to distinguish the Al-O environment. However, it should be notated that the occurrence of polyhedrons influence the position of the vibrations of Qi units as shown in Fig. 18. Comparing the results with the data summarized in42,43,44 with respect to iron phosphates, it can be detected that Al3+ shifts the Qi vibration toward higher frequencies compared to Fe3+. This may be useful in the case of materials containing iron and aluminum to differentiate Qi species connected with Al3+ and Fe3+ cations as in glasses. The position of the band related to the symmetric stretching modes of 3(P-ONB) in the structural unit of NaAlP2O7 is very close to the band in Q0 (α-cristobalite or α-berlinite). The position of this band for Q1 is usually higher than Q0 [19, 48]. The ionic nature of Na+ shifts the band toward lower values. A similar effect can be detected for Fe and Al, and iron, which is more ionic to oxygen than aluminum, also lowers the position of the band in Qi species19,42,45,46.
Another important observation is evidence of mid-intensity bands characteristic of phosphate rings vibrations in Al(PO3)3. The vibrations are located at higher frequencies next to the most intense band (Fig. 19). The bands characteristic to 6Q2- and 4Q2-rings are well visible and allow to distinguish between different Al(PO3)3 polymorphic forms.
Additionally, the main intense band for the ring structures is shifted toward the higher values, and the shift is the highest for the 4Q2-rings. The shift is probably related to the increase in the stiffness of bonds in ring structures. The rings are more rigid than the chains, and the smaller rings are more rigid than the larger ones. Therefore, the position moves to a higher frequency.
Conclusions
Theoretical Ramana and IR spectra of aluminum phosphate compounds containing Qi structural units were calculated from Q0 to Q3 and characteristic vibration modes were described. The selected compounds were synthesized, and the experimental spectra were compared with those of theoretical. It was evidenced by the good agreement between the theoretical and experimental results. The best convergence was obtained when the calculated Ramana and IR spectra were shifted in the range of + 20–+ 40 cm−1 without applying any scaling factor.
It was evidenced that the Raman spectra are characterized by the presence of two characteristic bands in the mid-and high-frequency ranges. The mid-band is originating from P–O–P bridges, whereas the higher band is the result of P–O vibrations in Qi tetrahedrons. The position of the high-frequency band is correlated with the index i in the Qi species and can be used to predict the distribution of Qi units in materials.
In the case of the Raman spectra, symmetric vibrations are much more intense than asymmetric, whereas in the case of the IR the opposite effect is evidenced. The IR spectra are also dominated by two bands because of the vibrations of P-O-P and P-O in Qi units, similar to Raman.
For Al(PO3)3 and AlPO4 differences in Raman spectra related to different polymorphic forms were observed and described.
Materials and methods
Simulations
Calculations of Raman and IR spectra were conducted for the crystalline compounds presented in Table 8 using Quantum Espresso 6.4 software47. In the calculation procedure, the unit cell parameter was taken from the reference, and the positions of the atoms were optimized. The unit cell parameters were not optimized to decrease the calculation time, especially for the big unit cells. This approach may limit the accuracy of the results. Nevertheless, most of the predicted spectra are compared to the experimental or literature data to validate the calculation procedure.
The PWscf program included in the Quantum Espresso package was used to optimize positions and perform self-consistent field SCF calculations. This program is based on Density Functional Theory (DFT), a plane-wave basis set, and pseudopotentials. The local density approximation LDA and optimized norm-conserving Vanderbilt scalar relativistic pseudopotentials from the Pseudo Dojo project49,50 were used in the calculations. The cut-off energy for valence electrons plane-waves basis set and charge densities were 50 and 200 Ry, respectively. The Monkhorst–Pack k-point sampling scheme with a 3 × 3 × 3 mesh grid was used. Self-consistency and convergence of total energy for ionic minimization were set to 10–8 and 10–4 Ry, respectively. The results of the SCF calculations for optimized structures of crystalline compounds were used in Raman and IR spectra calculations. The calculations of Raman and IR spectra were performed using the PHonon program from the Quantum Espresso package which is based on density functional perturbation theory (DFPT). The k-point grid remained the same as in the previous calculations. The threshold for self-consistency was set at 10–12 Ry. The selected k-point mesh was sufficient to obtain satisfactory results and at the same time a decent calculation time. To better visualize IR and Raman theoretical spectrum, the envelopes were calculated by a script written in Python using SciPy library51.
Synthesis
Crystalline compounds included in Table 8 were synthesized, except o’-P2O5 and B-Al(PO3)3. Stoichiometric quantities of chemically pure NH4H2PO4, Al2O3, and Na2CO3 were used. The synthesis was conducted according to the following procedure. The starting NH4H2PO4 was decomposed into H3PO4 by heating to 200 °C in a Al2O3 crucible in an electric furnace. The H3PO4 obtained was kept at 200 °C for 2 h. The molten H3PO4 was thoroughly mixed with Al2O3 or/and Na2CO3. The resulting pastes were placed in an alumina combustion boat. The samples were sintered according to the temperatures in Table 9. Synthesis temperatures were selected according to4,11,28. Due to the high hygroscopicity of P2O5, it must be synthesized in tightly closed containers. Also, the measurement procedure using XRD, Raman, or IR spectroscopy must be performed in the absence of air25. The synthesis of o’-P2O5 has been ongoing for several weeks. Due to these difficulties, it was decided to abandon the synthesis of o’-P2O5. B-Al(PO3)3 was not obtained from molten H3PO4 and Al2O3 at temperatures of 550 and 900 °C. A synthesis at 700 °C was also performed and a small amount of B-Al(PO3)3 is present in the sample but not enough to compare with the calculated spectra. This sample contains mainly A-Al(PO3)3 (Raman and IR spectra in the supplementary materials Fig. S2.1 and Fig. S2.2). Obtaining B-Al(PO3)3 from melting Al2O3 with HPO3 has been reported26. Also, in30 V. Bemmer et al. report only aluminum cyclohexaphosphate or/and A-Al(PO3)3 obtained from H3PO4 with various precursors (Al(OH)3, Al(NO3)3 or AlCl3) water solutions at temperatures 500 °C and 800 °C.
All of the steps of the synthesis were performed in an air atmosphere. The samples were gradually heated to the synthesis temperature for 5 h and then kept at the temperature for 8 h. Then were cooled to room temperature with the furnace. The obtained materials were visibly porous as a result of the release of water vapor during synthesis. The samples were then removed from the containers and crushed into smaller pieces. After berlinite synthesis, the part of Al2O3 did not react. Therefore, the sample was ground in an agate mortar and powder was pressed into a tablet using a hydraulic press. The pressed sample was sintered at 750 °C for 8 days.
The crystalline compositions of the samples were checked using XRD. Powder XRD measurements were carried out with a Philips X’Pert Pro diffractometer and Cu Kα1 radiation. The phase compositions of the obtained materials and the crystal structure parameters have been obtained using the Rietveld method using GSAS-II software52.
All Raman measurements were made using a Witec Alpha 300 M + Confocal Raman Imaging system with the application of a 50 × air objective (Zeiss, LD EC Epiplan-Neofluar, NA = 0.55). The spectrometer was equipped with an air-cooled solid-state laser operating at 488 nm, a CCD detector that was cooled to − 60 °C, and 600 grooves per mm of gratings. Raman spectra of each sample were collected with two scans and an integration time of 20 s.
Spectroscopic studies were carried out in middle infrared (MIR) regions (4000–400 cm−1) using a Fourier transformation spectrometer (FT-IR). Samples were prepared using tablet methods in KBr. Measurements were collected after 128 scans at a resolution of 4 cm−1.
Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
References
Wagh, A. S. Aluminum phosphate ceramics. In Chemically Bonded Phosphate Ceramics 141–155 (Elsevier, 2016). doi:https://doi.org/10.1016/B978-0-08-100380-0.00011-7.
Kominami, H., Matsuo, K. & Kera, Y. Crystallization and transformation of aluminum orthophosphates in organic solvent containing a small amount of water. J. Am. Ceram. Soc. 79, 2506–2508. https://doi.org/10.1111/j.1151-2916.1996.tb09008.x (1996).
Roy, A. K. & Sircar, N. R. A study of the clay-phosphoric acid system by thermal analysis. Trans. Indian Ceram. Soc. 41, 101–104. https://doi.org/10.1080/0371750X.1982.10822584 (1982).
Morris, J. H., Perkins, P. G., Rose, A. E. A. & Smith, W. E. The chemistry and binding properties of aluminium phosphates. Chem. Soc. Rev. 6, 173. https://doi.org/10.1039/cs9770600173 (1977).
Zhuravleva, P. L., Kitaeva, N. S., Shiryakina, Y. M. & Novikova, A. A. Study of thermal transformations of aluminum phosphate binder and composites on its basis with various fillers. Russ. J. Appl. Chem. 89, 367–373. https://doi.org/10.1134/S1070427216030046 (2016).
Bian, D. & Zhao, Y. Preparation and corrosion mechanism of graphene-reinforced chemically bonded phosphate ceramics. J. Sol–Gel Sci. Technol. 80, 30–37. https://doi.org/10.1007/s10971-016-4061-9 (2016).
Chiou, J.-M. & Chung, D. D. L. Improvement of the temperature resistance of aluminium-matrix composites using an acid phosphate binder. J. Mater. Sci. 28, 1435–1446. https://doi.org/10.1007/BF00363335 (1993).
Langlet, M., Saltzberg, M. & Shannon, R. D. Aluminium metaphosphate glass-ceramics. J. Mater. Sci. 27, 972–982. https://doi.org/10.1007/BF01197650 (1992).
Wang, M. et al. Effect of Al(PO3)3 content on physical, chemical and optical properties of fluorophosphate glasses for 2 μm application. Mater. Chem. Phys. 114, 295–299. https://doi.org/10.1016/j.matchemphys.2008.09.014 (2009).
Gan, F., Jiang, Y. & Jiang, F. Formation and structure of Al(PO3)3-containing fluorophosphate glass. J. Non. Cryst. Solids 52, 263–273. https://doi.org/10.1016/0022-3093(82)90301-5 (1982).
Ben Taher, Y., Hajji, R., Oueslati, A. & Gargouri, M. Infra-red, NMR spectroscopy and transport properties of diphosphate NaAlP2O7. J. Clust. Sci. 26, 1279–1294. https://doi.org/10.1007/s10876-014-0812-3 (2015).
Taher, Y. B., Oueslati, A., Khirouni, K. & Gargouri, M. Impedance spectroscopy and conduction mechanism of LiAlP2O7 material. Mater. Res. Bull. 78, 148–157. https://doi.org/10.1016/j.materresbull.2016.02.033 (2016).
Ben Taher, Y., Oueslati, A. & Gargouri, M. ac conductivity and NSPT model conduction of KAlP2O7 compound. Ionics (Kiel) 21, 1321–1332. https://doi.org/10.1007/s11581-014-1288-8 (2015).
Zhu, J. et al. Synthesis and red emitting properties of NaAlP2O7:Pr3+ polycrystal for blue chip-excited WLEDS. Results Phys. 12, 771–775. https://doi.org/10.1016/j.rinp.2018.12.047 (2019).
Chen, X., Lv, F., Ma, Y. & Zhang, Y. Preparation and spectroscopic investigation of novel NaAlP2O7:Eu2+ phosphors for white LEDs. J. Alloys Compd. 680, 20–25. https://doi.org/10.1016/j.jallcom.2016.04.125 (2016).
Hao, Y. et al. Highly porous aluminophosphates with unique three dimensional open framework structures from mild hydrothermal syntheses. CrystEngComm 22, 3070–3078. https://doi.org/10.1039/d0ce00075b (2020).
Komura, K., Aoki, H., Tanaka, K. & Ikeda, T. GAM-3: A zeolite formed from AlPO4-5: Via multistep structural changes. Chem. Commun. 56, 14901–14904. https://doi.org/10.1039/d0cc06086k (2020).
Parise, J. B. et al. Characterization of Se-Loaded Molecular Sieves A, X, Y, AIPO-5, and Mordenite. Inorg. Chem. 27, 221–228. https://doi.org/10.1021/ic00275a002 (1988).
Stoch, P., Stoch, A., Ciecinska, M., Krakowiak, I. & Sitarz, M. Structure of phosphate and iron-phosphate glasses by DFT calculations and FTIR/Raman spectroscopy. J. Non. Cryst. Solids 450, 48–60. https://doi.org/10.1016/j.jnoncrysol.2016.07.027 (2016).
Acelas, N. Y., Mejia, S. M., Mondragón, F. & Flórez, E. Density functional theory characterization of phosphate and sulfate adsorption on Fe-(hydr)oxide: Reactivity, pH effect, estimation of Gibbs free energies, and topological analysis of hydrogen bonds. Comput. Theor. Chem. 1005, 16–24. https://doi.org/10.1016/j.comptc.2012.11.002 (2013).
Stoch, A., Maurin, J., Kulawik, J. & Stoch, P. Structural properties of multiferroic 0.5BiFeO3–0.5Pb(Fe0.5Nb0.5)O3 solid solution. J. Eur. Ceram. Soc. 37, 1467–1476. https://doi.org/10.1016/j.jeurceramsoc.2016.11.029 (2017).
Rice, C. et al. Raman-scattering measurements and first-principles calculations of strain-induced phonon shifts in monolayer MoS2. Phys. Rev. B Condens. Matter Mater. Phys. 87, 1–5. https://doi.org/10.1103/PhysRevB.87.081307 (2013).
Janzen, B. M., Gillen, R., Galazka, Z., Maultzsch, J. & Wagner, M. R. First and second order Raman spectroscopy of monoclinic β-Ga2O3. Phy. Rev. 6, 054601 (2022).
Brow, R. K. Review: the structure of simple phosphate glasses. J. Non-Cryst. Solids 263, 1–28. https://doi.org/10.1016/S0022-3093(99)00620-1 (2000).
Stachel, D., Svoboda, I. & Fuess, H. Phosphorus pentoxide at 233 K. Acta Crystallogr. Sect. C Cryst. Struct. Commun. 51, 1049–1050. https://doi.org/10.1107/S0108270194012126 (1995).
Van der Meer, H. The crystal structure of a monoclinic form of aluminium metaphosphate, Al(PO3)3. Acta Crystallogr. B 32, 2423–2426. https://doi.org/10.1107/S0567740876007899 (1976).
Pauling, L. & Sherman, J. The crystal structure of aluminum metaphosphate, Al(PO3)3. Zeitschrift für Krist. - Cryst. Mater. 96, 481–487. https://doi.org/10.1524/zkri.1937.96.1.481 (1937).
Oudahmane, A., Mbarek, A., El-Ghozzi, M. & Avignant, D. Aluminium cyclohexaphosphate. Acta Crystallogr. Sect. E Struct. Reports https://doi.org/10.1107/S1600536810005374 (2010).
Vippola, M. et al. Structural characterization of aluminum phosphate binder. J. Am. Ceram. Soc. 83, 1834–1836. https://doi.org/10.1111/j.1151-2916.2000.tb01477.x (2000).
Bemmer, V. et al. Rationalization of the X-ray photoelectron spectroscopy of aluminium phosphates synthesized from different precursors. RSC Adv. 10, 8444–8452. https://doi.org/10.1039/c9ra08738a (2020).
Achary, S. N., Jayakumar, O. D., Tyagi, A. K. & Kulshresththa, S. K. Preparation, phase transition and thermal expansion studies on low-cristobalite type Al1−xGaxPO4 (x=0.0, 0.20, 0.50, 0.80 and 1.00). J. Solid State Chem. 176, 37–46. https://doi.org/10.1016/S0022-4596(03)00341-4 (2003).
Hummel, F. A. Properties of some substances isostructural with silica. J. Am. Ceram. Soc. 32, 320–326. https://doi.org/10.1111/j.1151-2916.1949.tb18905.x (1949).
Muraoka, Y. & Kihara, K. The temperature dependence of the crystal structure of berlinite, a quartz-type form of AlPO4. Phys. Chem. Miner. 24, 243–253. https://doi.org/10.1007/s002690050036 (1997).
Graetsch, H. A. High-temperature phase transitions and intermediate incommensurate modulation of the tridymite form of AlPO4. Zeitschrift fur Krist. 222, 226–233. https://doi.org/10.1524/zkri.2007.222.5.226 (2007).
Graetsch, H. Two forms of aluminium phosphate tridymite from X-ray powder data. Acta Crystallogr Sect. C Cryst. Struct. Commun. 56, 401–403. https://doi.org/10.1107/S0108270199015164 (2000).
Nicola, J. H. & Scott, J. F. Raman study of the α-β cristobalite phase transition in AlPO4. Phys. Rev. B 18, 1972 (1978).
Rokita, M., Handke, M. & Mozgawa, W. The AIPO4 polymorphs structure in the light of Raman and IR spectroscopy studies. J. Mol. Struct. 555, 351–356. https://doi.org/10.1016/S0022-2860(00)00620-7 (2000).
Westbrook, J. H. Temperature dependence of strength and brittleness of some quartz structures. J. Am. Ceram. Soc. 41, 433–440. https://doi.org/10.1111/j.1151-2916.1958.tb12891.x (1958).
Beck, W. R. Crystallographic inversions of the aluminum orthophosphate polymorphs and their relation to those of silica. J. Am. Ceram. Soc. 32, 147–151. https://doi.org/10.1111/j.1151-2916.1949.tb18940.x (1949).
Ng, H. N. & Calvo, C. X-ray study of the twinning and phase transformation of phosphocristobalite (AlPO4). Can. J. Phys. 55, 677–683. https://doi.org/10.1139/p77-095 (1977).
Saidi, M., Coffy, G. & Sibieude, F. Les systemes binaires AlPO4−M3PO4 (M=Li, Na, K). J. Therm. Anal. 44, 15–23. https://doi.org/10.1007/BF02547129 (1995).
Zhang, L. & Brow, R. K. A Raman study of iron-phosphate crystalline compounds and glasses. J. Am. Ceram. Soc. 94, 3123–3130. https://doi.org/10.1111/j.1551-2916.2011.04486.x (2011).
Okada, S. et al. Cathode properties of amorphous and crystalline FePO4. J. Power Sources 146, 570–574. https://doi.org/10.1016/j.jpowsour.2005.03.200 (2005).
Rojo, J. M., Mesa, J. L., Lezama, L. & Rojo, T. Magnetic properties of the Fe(PO3)3 metaphosphate. J. Solid State Chem. 145, 629–633. https://doi.org/10.1006/jssc.1999.8262 (1999).
Stoch, P. et al. Influence of aluminum on structural properties of iron-polyphosphate glasses. Ceram. Int. 46, 19146–19157. https://doi.org/10.1016/j.ceramint.2020.04.250 (2020).
Stoch, P. et al. Structural properties of iron-phosphate glasses: spectroscopic studies and ab initio simulations. Phys. Chem. Chem. Phys. 16, 19917–19927. https://doi.org/10.1039/C4CP03113J (2014).
Giannozzi, P. et al. QUANTUM ESPRESSO: A modular and open-source software project for quantum simulations of materials. J. Phys. Condens. Matter https://doi.org/10.1088/0953-8984/21/39/395502 (2009).
Alkemper, J., Paulus, H. & Fuei, H. Crystal structure of aluminum sodium pyrophosphate, NaAlP2O7. Zeitschrift für Krist. - Cryst. Mater. 209, 616–616. https://doi.org/10.1524/zkri.1994.209.7.616 (1994).
van Setten, M. J. et al. The PSEUDODOJO: Training and grading a 85 element optimized norm-conserving pseudopotential table. Comput. Phys. Commun. 226, 39–54. https://doi.org/10.1016/j.cpc.2018.01.012 (2018).
Hamann, D. R. Optimized norm-conserving Vanderbilt pseudopotentials. Phys Rev. B - Condens. Matter Mater. Phys. 88, 1–10. https://doi.org/10.1103/PhysRevB.88.085117 (2013).
Virtanen, P. et al. SciPy 1.0: Fundamental algorithms for scientific computing in Python. Nat. Methods 17, 261–272. https://doi.org/10.1038/s41592-019-0686-2 (2020).
Toby, B. H. & Von Dreele, R. B. GSAS-II: The genesis of a modern open-source all purpose crystallography software package. J. Appl. Crystallogr. 46, 544–549. https://doi.org/10.1107/S0021889813003531 (2013).
Acknowledgements
This research was partially funded by the National Science Center of Poland, grant number 2017/27/B/ST8/01477 and by the AGH-UST Initiative for Excellence Research University, Action 4, “Innovative glass-ceramic materials for the immobilization of radioactive and hazardous waste”. PG has been partly supported by the EU Project POWR.03.02.00-00-I004/16. The calculations were conducted thanks to PL-Grid Infrastructure.
Author information
Authors and Affiliations
Contributions
All authors reviewed the manuscript. B.H. and P.G. conducted research. P.G. performed ab-initio calculations, prepared figures and synthesized the samples. P.S. and P.G. wrote the main manuscript text and analyzed data.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Goj, P., Handke, B. & Stoch, P. Vibrational characteristics of aluminum–phosphate compounds by an experimental and theoretical approach. Sci Rep 12, 17495 (2022). https://doi.org/10.1038/s41598-022-22432-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-022-22432-5
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.